The long term goal of this project is to elucidate the molecular mechanism for liver cancer development. Toward this end, we developed and characterized a mouse model that carries deletion of a tumor suppressor that is often mutated (up to 44%) in human liver cancer. These mice, lacking the tumor suppressor PTEN (phosphatase and tensin homologue deleted on chromosome 10), displayed the progression from fatty liver disease to fibrosis to liver cancer and lung metastasis that has been observed in most human patients. Our preliminary data showed that a massive expansion of cells with progenitor cell properties occurs prior to tumor development in the Pten null model. We isolated these progenitor cells at the pre-malignant stage and found that they are tumor initiating cells (TICs) using xenograft models. In addition to isolating these cells, we have also begun to investigate the mechanism that lead to their growth and transformation. AKT is the best characterized downstream effector molecule of PTEN signaling. We found unexpectedly that loss of AKT2, known for its role in metabolic regulation but not AKT1, known for its role in cell growth/survival inhibits the development of tumors in the Pten null mice. A dramatic reduction in the progenitor cell activation morphology accompanied a significant delay in liver cancer development when Akt2 is simultaneously deleted with Pten. Our analysis shows that injury occurring with fatty liver is necessary for tumor development in the Pten null mice, whereas AKT2 attenuates fatty liver and injury. In this proposal, we plan to investigate the molecular and cellular mechanisms underlying the PTEN loss/AKT2 activation induced tumorigenesis. In the first aim, we will analyze the role of the AKT2 regulated lipogenesis on tumorigenesis and TIC cell activation. We will use dietary means to manipulate liver steatosis development and address the role of lipogenesis in tumorigenesis when PTEN or PTEN/AKT2 is lost. In the second aim, we will determine the effects of AKT2 on the transformation of the progenitor cells. We will overexpress myr-AKT2 and knockdown AKT2 in the unique Pten null and Pten/Akt2 double null liver progenitor cell lines that we have established and determine the effects of these manipulations on the abilities of these cells to form colonies, graft tumors and proliferate. We will additionally determine the molecular mechanisms (both canonical PI3K/AKT signaling and the unique metabolic signaling regulated by AKT2) that may underlie such a phenotype. In the last aim, we will determine the role of AKT2 in the upregulation of Wnt/?-catenin signaling observed with Pten deletion. We have found that Wnt/?-catenin, a pathway known to regulate progenitor cell activity, is activated in the Pten null mice. We hypothesize that AKT2 may antagonize the effect of PTEN on progenitor cell activation by controlling Wnt signaling. This may occur either through establishing a fatty liver/injury niche to induce Wnt expression or through directly regulating b-catenin activity within TICs. We will use both in TIC culture and in vivo approaches to address this aim. Together, these three aims will establish the causal effect of AKT2 activation on liver carcinogenesis in vivo and in vitro and the underlying mechanisms for such effect.